Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/maksimzayats/acodex/agents-mdgit clone --depth 1 https://github.com/maksimzayats/acodexWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/maksimzayats/acodex/agents-md)<a href="https://agentmods.dev/instructions/maksimzayats/acodex/agents-md"><img src="https://agentmods.dev/badge/instructions/maksimzayats/acodex/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00400 | $0.00400 |
| Opus 5 | $0.00200 | $0.00200 |
| Sonnet 5 | $0.00080 | $0.00080 |
| Haiku 4.5 | $0.00040 | $0.00040 |
Grade A, and why
acodex AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Agent Instructions
- Keep new features split by responsibility: CLI presentation in
cli, HTTP transport inhttp, Codex desktop bridging incore/codex_app, MCP client logic incore/mcp_tools.py, and dependency wiring inioc. - Use dependency injection for shared services and I/O boundaries. Register reusable collaborators in the container instead of constructing them ad hoc.
- Keep dependency wiring behind process entrypoints. Do not import
iocfrom feature modules, and do not makecli,http, orcoredepend on each other across the enforced import-linter boundaries. - Prefer focused classes or dataclasses for main feature logic, especially when dependencies, state, lifecycle, or test seams are involved. Keep standalone functions for small pure helpers and simple transformations.
- Keep package
__init__.pyfiles as public facades only. Do not add private backward-compatibility wrappers,compat.pyshims, or aliases just for old tests; update tests and callers to use the real class/module API instead. - Keep production code WPS-clean without production per-file ignores. Plain service classes should be slotted; Pydantic models/settings and exceptions are the main slotscheck exceptions.
- Treat Ruff, WPS/Flake8, mypy, import-linter, pyright, pyrefly, slotscheck,
pytest with 100% coverage, and
prek run --all-filesas required gates. Preserve these gates when changing tooling. - Exclude
references/andtmp/from lint, type, spelling, package, and test assumptions. They are local references only. - Update tests and docs with behavior changes. Keep user-facing content in
README.md, contributor workflow inCONTRIBUTING.md, and agent-only rules in this file.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 28 lines · 400 tokens per session scan A 649cd80635bd
acodex AGENTS.md is an instructions file published in the GitHub repository maksimzayats/acodex (57 stars, last pushed yesterday), licensed MIT. It adds 400 tokens to every session, about $0.0020 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
arcvellum AGENTS.md
AGENTS.md instructions for o-1717986918/arcvellum, covering agents, product boundary, development entry, worker constitution and verification.
atifact AGENTS.md
Instructions for waldekmastykarz/atifact, covering agents and atif spec version.
codexcw AGENTS.md
AGENTS.md instructions for c3-oss/codexcw, covering agents, project shape, build, test, develop, coding style and commits and prs.
dsh-codex-app-server AGENTS.md
Instructions for LyleMi/dsh-codex-app-server, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
dexgram AGENTS.md
Instructions for yashau/dexgram, covering agents.md, project overview, development commands, windows resource and binary version and runtime notes.
codexcw CLAUDE.md
Claude Code instructions for c3-oss/codexcw: See AGENTS.md for the canonical guidance. Conventions, layout, commands, hooks, and release trains all live there.